A method for real-time fusion of multi-sensor information of unmanned vehicle

• Main Authors: WANG YUZE, QIAO JIALE, ZHANG JINDONG
• Format: Patent
• Language: Chinese; English
• Published: 26.03.2019
• Subjects: CALCULATING; COMPUTING; COUNTING; DATA PROCESSING SYSTEMS OR METHODS, SPECIALLY ADAPTED FORADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL, SUPERVISORYOR FORECASTING PURPOSES; ELECTRIC DIGITAL DATA PROCESSING; PHYSICS; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE,COMMERCIAL, FINANCIAL, MANAGERIAL, SUPERVISORY OR FORECASTINGPURPOSES, NOT OTHERWISE PROVIDED FOR

The invention discloses a method for real-time fusion of multi-sensor information of an unmanned vehicle, which comprises the following steps: determining a priori probability of each sensor; Determining an initial mass function; Calculating the correlation degree; According to the correlation degree, the trust set, fuzzy set and risk set are obtained. Computing the risk function of trust set, fuzzy set and risk set; Calculating a posterior probability; Comparing the posterior probability of each sensor with a preset risk threshold; correcting The initial mass function of each sensor corresponding to each target information according to the following equation; Conducting D-S multiplication fusion, acquiring the fusion probability of each target information after the unmanned vehicle sensorsystem is recognized. A priori factor for measuring uncertainty is introduced to carry out risk assessment by using data association distribution, and the quantization treatment of uncertainty can effectively avoid the influ